Web-based practical privacy-preserving distributed image storage for financial services in cloud computing

With the development of multimedia technology and applications in the financial services industry, a large amount of multimedia image data related to financial services has been generated. For local users with limited resources including enterprises and individuals, it is a better choice to make full use of the rich storage capacity of the cloud. However, due to outsourcing the financial services image data to store on the cloud server, local users lose their direct control ability, which threatens the privacy of users. Although data security storage has attracted great attention from researchers, the protection of massive amounts of data, especially for financial services images, is brought to a new level with the advent of the era of artificial intelligence. In this paper, we focus on this issue and propose a lightweight data storage approach for privacy-preserving financial services images in the cloud. First, we use the EfficientDet neural network model to identify a large number of financial services images with sensitive data and classify the images. Then we perform different operations on the two types of images based on our proposed privacy-preserving financial services image storage architecture. Specifically, for privacy-sensitive financial services images, we use the privacy-preserving scheme we proposed to store the privacy parts on different servers, and directly outsource the storage of the remaining images without sensitive information. The experimental results show that our scheme achieves 2x-5x lower than some popular image encryption schemes on time consumption, and simultaneously protects the security of images. In particular, our scheme shows more excellent performance when it comes to storing a large number of financial services multimedia images.